Does physiological β cell turnover initiate autoimmune diabetes in
the regional lymph nodes?
Michal Pearl-Yafea, Svetlana Iskovicha, Ayelet Kaminitza, Jerry Steinb,
Isaac Yanivc, Nadir Askenasya,⁎
aFrankel Laboratory, Center for Stem Cell Research, Schneider Children's Medical Center of Israel, Israel
bBone Marrow Transplantation Unit, Schneider Children's Medical Center of Israel, Israel
cDepartment of Pediatric Hematology-Oncology, Schneider Children's Medical Center of Israel, Israel
Received 18 March 2006; accepted 4 April 2006
Available online 17 April 2006
The initial immune process that triggers autoimmune β cell destruction in type 1 diabetes is not fully understood. In early
infancy there is an increased β cell turnover. Recurrent exposure of tissue-specific antigens could lead to primary sensitization
of immune cells in the draining lymph nodes of the pancreas. An initial immune injury to the β cells can be inflicted by several
cell types, primarily macrophages and T cells. Subsequently, infiltrating macrophages transfer antigens exposed by apoptotic β
cells to the draining lymph nodes, where antigen presenting cells process and amplify a secondary immune reaction. Antigen
presenting cells evolve as dual players in the activation and suppression of the autoimmune reaction in the draining lymph
nodes. We propose a scenario where destructive insulitis is caused by recurrent exposure of specific antigens due to the
physiological turnover of β cells. This sensitization initiates the evolution of reactive clones that remain silent in the regional
lymph nodes, where they succeed to evade regulatory clonal deletion.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Autoimmune diabetes; Pancreatic lymph nodes; β cell turnover; Regulatory antigen presenting cells
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Take-home messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
β cell apoptosis and immune homeostasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Physiological susceptibility to anti-islet immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The regional lymph node “playground”. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Therapeutic immunomodulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Autoimmunity Reviews 5 (2006) 338–343
⁎Corresponding author. Frankel Laboratory, Center for Stem Cell Research, Schneider Children's Medical Center of Israel, 14 Kaplan Street,
Petach Tikva, 49202, Israel. Tel.: +972 3921 3954; fax: +972 3921 4156.
E-mail address: email@example.com (N. Askenasy).
1568-9972/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
The pathological process that results in diabetes has
two main stages: onset of autoimmunity and inflamma-
tory insulitis. The events that lead to the eruption of the
immune reaction against the β cell antigens remain elu-
cellular-molecular basis. One reason is our partial
understanding of the functional immune system. For
example, a basic question that arises in the context of
autoimmunity is the frequency at which autoreactive
clones evolve. Sensitization against self antigens may be
a common event that is successfully suppressed on a
regular basis in physiological conditions, or it may be a
sporadic, singular event that is difficult to control be-
apparently arises either from the escape of an autoim-
mune clone, or from incidental damage to the islets.
The events that follow are somewhat clearer. Upon
injury, the β cells expose additional antigens, which
become the targets for subsequent immune reactivity
against the islets. These antigens, including insulin epi-
topes, may be shed from the β cells or expressed on the
cell surface, further boosting the autoimmune process.
An immune reaction is processed and amplified in the
regional lymph nodes, where antigen-presenting cells
(APC) not only sensitize and activate islet-cytotoxic T
cells (TC), but also initiate regulatory mechanisms. Cur-
rent research focuses on cells with immunomodulatory
potential that reside in the lymphatic nodes draining the
pancreas, to gain further insights into the process of islet
inflammation and to test their therapeutic potential.
The high prevalence of autoimmune insulitis may
suggest that particular characteristics of the excretory
tissues make them a preferred target of the immune sys-
these tissues particularly exposed to immune reactivity,
or is their immune surveillance more difficult? Regard-
ing the first possibility, one may speculate that excretory
endocrine tissues, including islets and thyroid, often
present the secreted molecules in close spatial proximity
to structural membranal proteins, which become targets
cause involves certain characteristics of the lymphatic
drainage of these tissues that render them susceptible to
immune attack. The draining lymph nodes of the pan-
creas lie in proximity to the potent and highly-reactive
mesenteric lymphatic system that defends against po-
tential intestinal intrusions (Fig. 1). The lymphatics of
the thyroid are juxtaposed on another potent lymphatic
system, which guards the upper respiratory tract. It
would be interesting to know whether these lymphatic
systems have particular physical or functional character-
istics in subjects affected by autoimmune reactions
against these tissues. For example, functional immune
elements may erroneously interfere and perturb the
subtle immune homeostasis in the regional lymph nodes
of the pancreatic tissue. Is it the proactive mode of
operation of the intestinal lymphatic system that is
permissive tothe development of autoimmune clones, or
is it a functional promiscuity that makes possible the
evasion of anti-islet reactive cells? It is also possible that
particular characteristics of the subsets of autoimmune
clones in these tissues help them to avoid immune sur-
veillance. Whatever the cause of the deregulation in im-
mune homeostasis that underlies autoimmunity, it is
evident from animal models of diabetes that the acti-
vation of a single T cell clone against a single antigen
expressed by the β cells is sufficient for the evolution of
2. β cell apoptosis and immune homeostasis
β cells in the inflamed islets die by apoptosis, as
demonstrated by spontaneous and accelerated diabetes
models of transgenic mice engineered to express anti-
genic targets in β cells [1–3]. Both the onset and the
propagation of the anti-islet reactivity should be viewed
in the context of the physiological mode of action of the
immune system, as presently known to us. Immune ho-
meostasis is sustained in ever changing environments.
Activation is frequently a vigorous event and suscepti-
bility to activation induced cell death (AICD) is an in-
trinsic feature of immune cell stimulation [4–6]. This
expression upon activation of immune cells [5,6,
reviewed in Ref. 7]. It is important to mention that all
lymphocytes (including CD4+and CD8+T, B and NK
cells), dendritic cells (DC), monocytes, macrophages,
and neutrophils are subjected to Fas/FasL-regulated
immune homeostasis [8,9]. The interaction of Fas with
FasL triggers autocrine and paracrine apoptosis, control-
ling the size of the pool of antigen-sensitized T cell
clones and regulates the magnitude of the immune re-
sponse [6,7]. Thus, AICD preserves self-antigen toler-
ance and prevents autoimmunity [7–9]. In some
situations the elimination of reactive cells via AICD is
not immediate. T cells are believed to become sensitive
to Fas/FasL-mediated apoptosis after several rounds of
antigenic challenge , and additional time elapses
until the cells acquire sensitivity to Fas-mediated apo-
in the process of antigen recognition and presentation,
339M. Pearl-Yafe et al. / Autoimmunity Reviews 5 (2006) 338–343
inflict direct damage to the target tissue. In addition,
some of the effector cells may migrate and reside in non-
lymphoid tissues, and their emergence occurs after ex-
posure to a recurrent specific antigenic challenge .
3. Physiological susceptibility to anti-islet immunity
Within this mode of immune activation, the pancre-
atic β cells may become targets of autoimmunity
because of the recurring physiological exposure of
tissue-specific antigens. Initially, it was thought that β
cells are terminally differentiated cells, which persist (in
health) throughout most of the adult life. More recent
evidence, however, points to a dynamic turnover of the
β cells, which have a finite life span and are renewed by
tissue progenitors [13,14]. Several studies reported the
death of β cells in newborn mice, culminating at 2–3
weeks of age [13,15]. In NOD mice this time point
coincides with the earliest onset of autoimmunity .
In humans, a similar phenomenon has been proposed to
underlie the recovery from hyperinsulinemia of infancy
. This wave of early β cell death may play a role in
the sensitization of the immune system, wherein profes-
sional APC process the β cell antigens in the draining
lymph nodes [13,18]. Islet remodeling persists into ma-
ture life, with continuous adjustment to the increasing
insulin requirements, though there is no apparent in-
crease in the density of islets in mature mice . This
Fig. 1. Initiation and amplification of the anti-β cell autoimmune reaction: the role of regional lymph nodes. The initial injury to the islets (?) may be
caused by macrophages (MP), neutrophils (PMN), autoreactive CD4+and CD8+Tcells (TC) that escaped clonal deletion. The exposed islet-specific
antigens are carried by macrophages to the regional lymph nodes where they are processed by dendritic cells (DC). The presentation of β cell-specific
antigens by MP and/or DC to CD4+Tcells in association with major histocompatibility complex antigens results in the development of cytotoxic T
cells (CTC) of both CD4+and CD8+phenotypes. The killing activity of sensitized CTC in the peripheral pancreatic lymph nodes is amplified by
cytokines in the inflammatory environment in the pancreatic islets. When the β cell mass is reduced to 10–30% the islet-inflammatory condition
(insulitis) converts to overt diabetes. The pancreatic lymph nodes also host DC-mediated activation of regulatory T cells, which act to suppress
inflammation. Insert: the lymphatic drainage of the pancreas, the pancreato-splenic chains of lymph nodes, are connected to the pyloric, mesenteric,
hepatic, superior mesenteric (SM) and celiac lymphatic trunks.
340M. Pearl-Yafe et al. / Autoimmunity Reviews 5 (2006) 338–343
process involves both adaptive growth in β cell mass
and apoptotic death (in the absence of anti-islet immuni-
ty) . In aged mice, there is a marked decline in the
turnover of β cells [13,19,20]. However, there are sig-
nificant differences in the specific rate estimates among
various studies, probably because of the inclusion of
non-excreting apoptotic cells which may have become
senescent prior to death. It is evident that β cell mitosis
is continuously stimulated by low doses of streptozo-
tocin . The measurement of β cell turnover is also
limited by the very low incidence of dead cells in the
islets . Nevertheless, the death rates determined by
examination at a single time point reflect primarily the
efficiency of dead cell clearance. Effective clearance of
dead β cells by macrophages to the draining lymph
nodes  may be one of the reasons of the high im-
munogenicity of this tissue. Thus, continuous β cell
death corresponds to the principle of repeated antigenic
sensitization, which may generate silent reactive clones
in the regional lymph nodes.
4. The regional lymph node “playground”
The pancreas is not an immune privileged site
[reviewed in Ref. 22]. Infiltrating macrophages carry
the antigens to the draining lymph nodes, for uptake,
processing, and presentation of the cognate molecules
by naïve T cells (TC) and professional APC [21,23].
Apoptotic and necrotic cells are potent stimulants of
APC, in particular dendritic cells (DC). In an elegant
experiment it was shown that excision of the pancreatic
draining lymph nodes from 3 week-old NOD mice pre-
vented the evolution of autoimmune diabetes . The
disease could be not prevented by lymph node excision
at 10 weeks (after onset of insulitis) and by splenectomy.
It is possible that in situ differentiation of pathogenic
evasion from regulatory immune homeostasis (Fig. 1).
The continuous inflow of antigens from the dead β cells
may provide the means of repeated antigenic challenge
of reactive, but silent, immune cells in the draining
lymph nodes. These reactive clones may become acti-
vated whenasecondaryhit(environmental,metabolic or
physiological) re-exposes the islet antigens. This mecha-
nism awaits experimental validation. Indirect evidence
supporting the role of the draining lymphatics as a hide-
cells in the spleens of diabetics, suggesting that the
reactive cells may indeed emerge from the lymph nodes
One of the major problems in identifying the reactive
T cells in autoimmune disorders in general and partic-
ularly in diabetes, is the difficult identification of the
specific target antigens. Analysis of the T cell receptor
(TCR) repertoire in islet-infiltrating and lymphatic T
cells revealed marked heterogeneity, suggesting sensiti-
zation against multiple antigens [26,27]. Evaluations of
non-obese diabetic (NOD) mice failed to determine
whether autoimmunity is directed against multiple anti-
gens, or whether β cell death exposes intracellular anti-
gens that serve as secondary molecular targets. Diabetic
antigens including Gad65 and insulin, were shown to be
involved in the evolution of insulitis in NOD mice
[28,29, reviewed in Ref. 30], and a recent study reported
the presence of Tcell clones reactive to insulin epitopes
in the regional lymph nodes of human diabetics .
Interference with insulin expression and tolerization to
proinsulin indeed affected but it did not abrogate diabe-
tes in mice . Thus, insulin may be a primary anti-
genic target of autoimmunity, or may become a target at
a later stage; in both cases it augments β cell death. In
more general terms, it would be potentially useful to
know whether diabetes is associated with a progressive
sensitization against secondary antigens after onset of
inflammatory insulitis [reviewed in Refs. 32,33].
5. Therapeutic immunomodulation
Localization and sequestration of the autoimmune
reactive clones, and the set of immune cells that partici-
pate in propagation of the reaction in the regional pan-
creatic lymph nodes impose difficulties on therapeutic
approaches. Currently efforts are directed towards ma-
nipulation of immune cells to achieve tolerogenic im-
munomodulation [reviewedinRefs.22,34,35].The anti-
inflammatory efficacy of DC isolated from local
pancreatic lymph nodes was markedly higher than that
of naïve and splenic DC, suggesting that these DC could
suppress autoimmunity by activation of regulatory
mechanisms [36–39]. Development of diabetes in
NOD mice could be prevented by injection of pancreatic
lymph node cells and interferon-γ (INFγ)-pulsed DC,
which lead to increased levels of Th2-type cytokines in
pancreas-infiltrating cells and reduced β cell apoptosis
[38, reviewed in Ref. 40]. Any attempt to treat auto-
immunity with non-specific, systemic, immunosuppres-
sive therapy, should consider the abrogation of the
regulatory activities of APC and T cells.
It is possible that the physiological turnover of β cells
341M. Pearl-Yafe et al. / Autoimmunity Reviews 5 (2006) 338–343
At this site, antigen presenting cells process the antigens
and sensitize Tcell clones, and at the same time, activate
with immunomodulatory potential that reside in lym-
phatic tissues draining the pancreas should be further
studied to determine their therapeutic potential. It is
conceivable that an initial injury leads to the exposure of
shed from the islets further sensitize autoreactive Tcells.
The difficulties we encounter in definition of the
autoimmune process in diabetes may be caused by
shifting targets of immunity in this disease.
This study was supported by grants from the Leah
and Edward M. Frankel Trust for Experimental Bone
Marrow Transplantation, the Daniel M. Soref Charitable
Trust and US-Israel BSF grant 2003276.
• Physiological turnover of β cells may participate in
exposure of specific antigens to the immune system.
• The major events in sensitization of the autoimmune
reaction take place in the pancreatic lymph nodes.
• The autoimmune reaction against β cells appears to
be directed against multiple antigens that evolve
along the course of the inflammation.
• After onset of the autoimmune reaction, the optimal
approach to arrest insulitis is the activation of
regulatory and suppressive antigen presenting cells
and T cells.
RC. DNA fragmentation is an early event in cytokine-induced
islet beta-cell destruction. Diabetologia 1994;37:733–8.
 Kurrer MO, Pakala SV, Hanson HL, Katz JD. β Cell apoptosis in
Tcell-mediated autoimmune diabetes. Proc Natl Acad Sci U S A
 Mauricio D, Mandrup-Poulsen T. Apoptosis and the patho-
genesis of IDDM: a question of life and death. Diabetes
 Cohen JJ, Duke RC. Apoptosis and programmed cell death in
immunity. Annu Rev Immunol 1992;10:267–93.
 Kabelitz D, Pohl T, Pechhold K. Activation-induced cell death
(apoptosis) of mature peripheral T lymphocytes. Immunol Today
 Russell JH. Activation-induced death of mature T cells in the
regulation of immune responses. Curr Opin Immunol 1995;7:
 Lenardo M, Chan KM, Hornung F, McFarland H, Siegel R,
Wang J, et al. Mature T lymphocyte apoptosis—immune
regulation in a dynamic and unpredictable antigenic environ-
ment. Annu Rev Immunol 1999;17:221–53.
 Van Parijs L, Abbas AK. Homeostasis and self-tolerance in the
immune system: turning lymphocytes off. Science 1998;280:
 Zhang HG, Su X, Liu D, Liu W, Yang P, Wang Z, et al. Induction
of specific T cell tolerance by Fas ligand-expressing antigen-
presenting cells. J Immunol 1999;162:1423–30.
 Inaba M, Kurasawa K, Mamura M, Kumano K, Saito Y, Iwamoto I.
Primed Tcells are more resistant to Fas-mediated activation-induced
cell death than naive T cells. J Immunol 1999;163:1315–20.
 O'Flaherty E, Ali S, Pettit SJ, Kirby JA. Examination of the
sensitivity of T cells to Fas ligation: induction of allospecific
apoptosis. Transplantation 1998;66:1067–73.
 Masopust D, Vezys V, Marzo AL, Lefrancois L. Preferential
localization of effector memory cells in nonlymphoid tissue.
 Finegood DT, Scaglia L, Bonner-Weir S. Dynamics of beta-cell
mass in the growing rat pancreas. Diabetes 1995;44:249–56.
 Dor Y, Brown J, Martinez OI, Melton DA. Adult pancreatic beta-
cells are formed by self-duplication rather than stem-cell
differentiation. Nature 2004;429:41–6.
 Trudeau JD, Dutz JP, Arany E, Hill DJ, Fieldus WE, Finegood
DT. Neonatal beta-cell apoptosis: a trigger for autoimmune
diabetes? Diabetes 2000;49:1–7.
 Andre I, Gonzalez A, Wang B, Katz J, Benoist C, Mathis D.
Checkpoints in the progression of autoimmune disease: lessons
from diabetes models. Proc Natl Acad Sci U S A 1996;93:2260–3.
 Kassem SA, Ariel I, Thornton PS, Scheimberg I, Glaser B. Beta-
cell proliferation and apoptosis in the developing normal human
pancreas and in hyperinsulinism of infancy. Diabetes 2000;49:
 Mathis D, Vence L, Benoist C. Beta-Cell death during
progression to diabetes. Nature 2001;414:792–8.
 Montanya E, Nacher V, Biarnes M, Soler J. Linear correlation
between beta-cell mass and body weight throughout the lifespan
in Lewis rats: role of beta-cell hyperplasia and hypertrophy.
 Teta M, Long SY, Wartschow LM, Rankin MM, Kushner JA.
Very slow turnover of β-cells in aged adult mice. Diabetes
 Katz JD, Wang B, Haskins K, Benoist C, Mathis D. Following a
diabetogenic T cell from genesis through pathogenesis. Cell
 Askenasy N, Yolcu ES, Isaac Yaniv I, Shirwan H. Induction of
tolerance using Fas ligand: a double-edged immunomodulator.
 Hoglund P, Mintern J, Waltzinger C, Heath W, Benoist C, Mathis
D. Initiation of autoimmune diabetes by developmentally
regulated presentation of islet cell antigens in the pancreatic
lymph nodes. J Exp Med 1999;189:331–9.
 Gagnerault MC, Luan JJ, Lotton C, Lepault F. Pancreatic lymph
nodes are required for priming of β cell reactive T cells in NOD
mice. J Exp Med 2002;196:369–77.
 Kent SC, Chen Y, Bregoli L, Clemmings SM, Kenyon NS,
Ricordi C, et al. Expanded T cells from pancreatic lymph nodes
of type 1 diabetic subjects recognize an insulin epitope. Nature
 Candeias S, Katz J, Benoist C, Mathis D, Haskins K. Islet-
specific T-cell clones from nonobese diabetic mice express
342 M. Pearl-Yafe et al. / Autoimmunity Reviews 5 (2006) 338–343
heterogeneous T-cell receptors. Proc Natl Acad Sci U S A Download full-text
 BakerF,Lee M,ChienY, DavisM.Restrictedislet-cell reactiveT
cell repertoire of early pancreatic islet infiltrates in NOD mice.
Proc Natl Acad Sci U S A 2002;99:9374–9.
 Tisch R, Yang XD, Singer SM, Liblau RS, Fugger L, McDevitt
HO. Immune response to glutamic acid decarboxylase
correlates with insulitis in non-obese diabetic mice. Nature
 Wegmann DR, Norbury-Glaser M, Daniel D. Insulin-specific T
cells are a predominant component of islet infiltrates in pre-
diabetic NOD mice. Eur J Immunol 1994;24:1853–7.
 Jasinski JM, Eisenbarth GS. Insulin as a primary autoantigen for
type 1A diabetes. Clin Dev Immunol 2005;12:181–6.
 Jaeckel E, Lipes MA, von Boehmer H. Recessive tolerance to
preproinsulin 2 reduces but does not abolish type 1 diabetes. Nat
 Bottino R, Lemarchand P, Trucco M, Giannoukakis N. Gene- and
cell-based therapeutics for type I diabetes mellitus. Gene Ther
 Pearl-Yafe M, Yolcu ES, Yaniv I, Stein J, Shirwan H, Askenasy
N. The dual role of Fas-ligand as an injury effector and defense
strategy in diabetes and islet transplantation. Bioessays 2006;28:
 Filippi C, Bresson D, Herrath M. Antigen-specific induction of
regulatory T cells for type 1 diabetes therapy. Int Rev Immunol
 Goudy KS, Tisch R. Immunotherapy for the prevention and
treatment of type 1 diabetes. Int Rev Immunol 2005;24:307–26.
 Clare-Salzler MJ, Brooks J, Chai A, Van Herle K, Anderson C.
Prevention of diabetesin nonobesediabeticmiceby dendriticcell
transfer. J Clin Invest 1992;90:741–8.
 Feili-Hariri M, Dong X, Alber SM, Watkins SC, Salter RD,
Morel PA. Immunotherapy of NOD mice with bone marrow-
derived dendritic cells. Diabetes 1999;48:2300–8.
 Shinomiya M, Fazle Akbar SM, Shinomiya H, Shinomiya H,
Onji M. Transfer of dendritic cells (DC) ex vivo stimulated with
interferon-gamma (IFNgamma) down-modulates autoimmune
diabetes in non-obese diabetic (NOD) mice. Clin Exp Immunol
 Papaccio G, Nicoletti F, Pisanti FA, Bendtzen K, Galdieri M.
Prevention ofspontaneousautoimmune diabetesin NODmiceby
transferring in vitro antigenpulsed syngeneic dendritic cells.
 Ludewig B, Odermatt B, Ochsenbein AF, Zinkernagel RM,
Hengartner H. Role of dendritic cells in the induction and
maintenance of autoimmune diseases. Immunol Rev 1999;169:
CD4 memory T cells survive and proliferate but fail to differentiate in the absence of CD40.
Correlations between anti-nucleosome antibodies and anti-dsDNA antibodies, C3, C4, and clinical activity in lupus
Secondary T cell responses are enhanced because of an expansion in numbers of antigen-specific (memory) cells.
Here, using major histocompatibility complex class II tetramers, Macleod M. et.al. (J Exp Med 2006; 203: 897-906)
have tracked peptide-specific endogenous (non-T cell receptor transgenic) CD4 memory T cells in normal and in
costimulation-deficient mice. CD4 memory T cells were detectable after immunization for more than 200 days,
wild-type dendritic cells survived in the absence of CD40 and proliferated when boosted with peptide (pulse
adjuvant) in a CD40-independent fashion. However, differentiation of the memory cells into cytokine-producing
effector cells did not occur in the absence of CD40. The data indicate that memory cells can be generated without
passing through the effector cell stage.
SLE patients. The prevalence of anti-nucleosome antibodies were higher than anti-dsDNA antibodies (52.3% vs
36.9%, respectively, p < 0.05). Similar results were obtained in 45 active SLE patients, 64.4% for anti-nucleosome
antibodies and 46.7% for anti-dsDNA antibodies. Of 34 patients lacking anti-dsDNA antibodies, 16 (47.1%) were
anti-nucleosome antibodies could be one of the earliest and most sensitive markers in diagnosis of SLE, particularly
343 M. Pearl-Yafe et al. / Autoimmunity Reviews 5 (2006) 338–343